Research On Prediction Of Internal Leakage Fault Of Hydraulic Motor Based On Simulated Experimental Platform

As one of the important system in modern aircraft,hydraulic system will directly affect the operation reliability of flight control system and landing gear system.As an important component of hydraulic system,once the hydraulic motor output power is reduced because of fault,the hydraulic motor could not drive the load to work normally,and that would lead to a fatal accident.The internal leakage fault of hydraulic motor is not easy to be detected when it occurs,so the leakage degree would gradually increase.In order to avoid accidents caused by hydraulic motor,the internal leakage fault of the hydraulic motor is studied in this paper.In order to study fault prediction for internal leakage of hydraulic motor,a fault simulation experimental platform is built,and the fault is injected in a non-destructive way.Before the physical platform is built,a hydraulic motor internal leakage fault simulation model with throttle valve was established in AMESim simulation software to obtain the simulation parameters of the hydraulic system under non-destructive fault injection.After that,the simulation model with leakage sub-model was also established to simulate the internal leakage fault of hydraulic motor caused by wear,and the simulation parameters of the hydraulic system state were also obtained.By comparing the state data of the two simulation models,the corresponding relationship between them is analyzed,and the feasibility and rationality of the non-destructive fault injection method is verified.Computer aided test technology is used in the design of hydraulic motor internal leakage fault simulation experimental platform,so not only the hydraulic system and its control system,but also the data acquisition system is designed.After the hydraulic circuit is designed,the selection of hydraulic components is completed according to the design conditions.Data acquisition card and sensor are selected according to the parameters to be monitored in the hydraulic circuit.By using LabVIEW software,the data acquisition,display and storage program of the simulation experimental platform is designed,so that the data collected by the sensor can be sent to the computer by data acquisition card for display and storage.The data acquisition process and the electromagnetic directional valve can be controlled by the control system,the hydraulic system can also be shut down when the pressure exceeds the limit.After the data are collected and stored in the computer,the abnormal data are removed through data preprocessing.In this paper,the fault prediction method based on fuzzy neural network is used to predict the internal leakage fault of hydraulic motor.After the prediction model is trained with different part of experimental data,the prediction results are obtained,and the accuracy of the prediction results is verified with the remaining data.After analyzing the prediction results,the influence of different number of samples used for model training on the accuracy of fault prediction is discussed.It is found that where the data fluctuation is large,the relative error is large.Therefore,after fitting the experimental data,the model is trained again in the same way and the prediction results are obtained,and the difference in prediction accuracy before and after fitting is analyzed.The results show that the more the data is used for model training,the higher the prediction accuracy is,and after data fitting,less data is needed for model training to achieve the same prediction accuracy.